Human aggression constitutes a complex a multidetermined act which results in physical or verbal assault to others, self, or objects. Among the various biological factors studied, the most consistent factor associated with impulsive (but not premeditated) aggression is a reduction in the activity of the central monoamine neurotransmitter serotonin (5-HT). 5-HT's role in the regulation of impulsive aggressive behavior is thought to arise from its role in neuronal inhibition, through which it raises the threshold for action in response to an outward (or inward) stimulus. 5-HT appears to act to restrain the individual from assaulting the object of stimulus. A role central 5-HT in the regulation of impulsive aggression and/or suicidal (i.e., of a violent and/or impulsive nature) behavior in humans is suggested by the association of alterations in indices of 5-HT activity (e.g., reductions in brain 5-HT/5-HIAA, lumbar CSF 5-HIAA, brain/platelet 5-HT Transporter binding, PRL response to 5-HT Challenge) in psychiatric, increases in brain/platelet 5-HT-2a receptor binding) patients with prominent histories of impulsive aggressive and suicidal behavior. During the current grant period, the applicant conducted an experimental clinical trial in which the selective 5-HT uptake inhibitor, Fluoxetine, was found to significantly reduce the frequency and severity of impulsive aggressive behaviors in a 12-week randomized, double-blind, placebo-controlled trial in personality disordered subjects with prominent histories of impulsive aggressive behavior. Subjects in this were found to have reduced PRL responses to d-Fenfluramine Challenge (PRL[d-FEN]). In addition, PRL[d-FEN] values appeared to correlate positively with anti-aggressive response to fluoxetine and to decrease after successful treatment with fluoxetine in responders. In this renewal application, the applicant proposes to: a) extend their clinical trial to include a 12-week Maintenance Treatment Phase and a 12-week Placebo-Controlled Discontinuation Phase in order to collect data regarding anti-aggressive efficacy beyond the initial 12 weeks of treatment and to determine the stability of remission in impulsive aggressive behaviors after 24 weeks of treatment with Fluoxetine; b) determine the relationship between Pre-Treatment and Post-Treatment indices of 5-HT activity (i.e., PRL[d-FEN] responses and Platelet 5-HT Transporter/5-HT-2a Receptor Binding) and anti-aggressive efficacy to Fluoxetine; and c) test the relative anti-aggressive efficacy of Fluoxetine in the three most prominent Personality Disorder subtypes occurring in DSM-IV Personality Disordered subjects with prominent histories of impulsive aggressive behavior.